Tudor Lapusan's Blog

Post info:

Visual interpretation of Decision Tree structure

In Machine Learning it’s important to understand why based on specific inputs (model hyperparameters, features or training set) your models generate some specific outputs (model performance measured by loss functions). My opinion is if we just measure the model performance we will don’t have the full picture of what’s happening behind, so we may end up luckily selecting the set of hyperparameters which we think generate the best model. Maybe the worst thing is that for the next ML project we

Read the full post

Post info:

Perfect fit : Apache Spark, Zeppelin and Docker

The goal of this article is to show how easy you can start working with Apache Spark using Apache Zeppelin and Docker. I played for the first time with docker when Cloudera announced the new quickstart option for trying Apache Hadoop with Cloudera. It was a really nice experience and I was surprised by docker characteristics. For me the most powerful characteristic was the ability to share containers between users using for exemple Docker Hub. Here is the official definition of containers

Read the full post